Abstract
The recent progress in geographical information systems, remote sensing (RS) and data analytics enables us to acquire and process large amount of Earth observation data. Convolutional neural networks (CNN) are being used frequently in classification of multi-dimensional images with high accuracy. In this paper, we test CNNs for the classification of hyperspectral RS data. Our proposed CNN is a multi-layered neural network architecture, which is tailored to classify objects based on pixel-wise spatial information using spectral bands of hyperspectral imagery (HSI). We use benchmark satellite imagery in four different HSI datasets for classification using the proposed architecture. Our results are compared with support vector machine (SVM) and extreme learning machine (ELM) algorithms, which are frequently used techniques of machine learning in RS data classification. Moreover, we also provide a comparison with the state-of-the-art CNN approaches, which have been used for HSI classification. Our results show improvements of up to 6% on average over SVM and ELM while up to 4% improvement is observed in comparison with two recently proposed CNN architectures for HSI classification accuracy. On the other hand, the processing time of our proposed CNN is also significantly lower.
Zusammenfassung
Pixelweise Klassifizierung von Hyperspektralszenen mit Convolutional Neural Networks. Der Fortschritt bei Geoinformationssystemen, Fernerkundung und Datenanalyse erlaubt uns die Gewinnung und Verarbeitung von umfangreichen Erdbeobachtungdaten. Convolutional Neural Networks (CNN) werden oft zur Klassifizierung von multidimensionalen hoch aufgelösten Bilddaten verwendet. In diesem Artikel untersuchen wir die Eignung von CNNs für die Klassifizierung von hyperspektralen Fernerkundungsdaten. Das von uns vorgeschlagene CNN besitzt die Struktur eines neuronalen Netzwerks mit mehreren Ebenen zur Objekt-Klassifizierung auf der Grundlage einer pixelweisen Auswertung der hyperspektralen Bilddaten. Zur Verifizierung unserer Klassifizierungsmethode benutzen wir vier verschiedene Datensätze, aufgenommen von Satellitenplattformen. Die Ergebnisse werden mit denen der Methoden Support Vector Machine (SVM) und Extreme Learning Machine (ELM), die beide bei automatischen Klassifizierungsverfahren der Fernerkundung weit verbreitet sind, verglichen. Darüber hinaus liefern wir einen Vergleich zu aktuellen Ansätzen der CNN. Unsere Ergebnisse zeigen eine Verbesserung der Klassifizierungsgenauigkeit von 6% gegenüber SVM und ELM sowie eine Verbesserung von 4% gegenüber kürzlich veröffentlichen CNN-Architekturen. Darüber hinaus ist unser Ansatz deutlich schneller.
Similar content being viewed by others
References
Abraham A (2005) Artificial neural networks. In: Sydenham PH, Thorn R (eds) Handbook of measuring system design. https://doi.org/10.1002/0471497398.mm421
Ba J, Mnih V, Kavukcuoglu K (2014) Multiple object recognition with visual attention. arXiv:1412.7755
Belgiu M, Drgu L (2016) Random forest in remote sensing: a review of applications and future directions. ISPRS J Photogramm Remote Sens 114:24–31
Bengio Y (2009) Learning deep architectures for AI. Found Trends Mach Learn 2(1):1–127
Blaschke T (2010) Object based image analysis for remote sensing. ISPRS J Photogramm Remote Sens 65:2–16
Bottou L (2012) Stochastic gradient descent tricks. In: Montavon G, Orr GB, Müller KR (eds) Neural networks: tricks of the trade. Lecture Notes in Computer Science. Springer, Berlin, vol 7700. pp 421–436
Dan C, Meier U, Schmidhuber J (2012) Multi-column deep neural networks for image classification. arXiv:1202.2745
Cao X, Zhou F, Xu L, Meng D, Xu Z, Paisley J (2017) Hyperspectral image segmentation with markov random fields and a convolutional neural network. arXiv:1705.00727
Chetlur S, Woolley C, Vandermersch P, Cohen J, Tran J, Catanzaro B et al (2014) CuDNN: efficient primitives for deep learning. arXiv:1410.0759
Dahl GE, Sainath TN, Hinton GE (2013) Improving deep neural networks for LVCSR using rectified linear units and dropout. In: Acoustics, speech and signal processing (ICASSP), 2013 IEEE international conference on, IEEE, pp 8609-8613
Guidici D, Clark ML (2017) One-dimensional convolutional neural network land-cover classification of multi-seasonal hyperspectral imagery in the San Francisco Bay Area, California. Remote Sens 9:629
Ham J, Chen Y, Crawford MM, Ghosh J (2005) Investigation of the random forest framework for classification of hyperspectral data. IEEE Trans Geosci Remote Sens 43(3):492–501
Hu W, Huang Y, Wei L, Zhang F, Li H (2015) Deep convolutional neural networks for hyperspectral image classification. J Sens 2015:258619. https://doi.org/10.1155/2015/258619
Huang GB, Zhu QY, Siew CK (2006) Extreme learning machine: theory and applications. Neurocomputing 70(1–3):489–501
Jia Y, Shelhamer E, Donahue J, Karayev S, Long J, Girshick R et al (2014) Caffe: convolutional architecture for fast feature embedding. In: Proceedings of the 22nd ACM international conference on multimedia. Florida, USA
Kampffmeyer M, Salberg A-B, Jenssen R (2016) Semantic segmentation of small objects and modeling of uncertainty in urban remote sensing images using deep convolutional neural networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition workshops, pp 1–9
Kasturi R, Goldgof D, Soundararajan P, Manohar V, Garofolo J, Bowers R et al (2009) Framework for performance evaluation of face, text, and vehicle detection and tracking in video: data, metrics, and protocol. IEEE Trans Pattern Anal Mach Intell 31:319–336
Kim Y (2014) Convolutional neural networks for sentence classification. arXiv:1408.5882
Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. In: Advances in neural information processing systems
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp 1097–1105
Landgrebe D (2002) Hyperspectral image data analysis. IEEE Signal Process Mag 19:17–28
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521:436–444
LeCun Y, Huang FJ, Bottou L (2004) Learning methods for generic object recognition with invariance to pose and lighting. In: Proceedings of the 2004 IEEE computer society conference on computer vision and pattern recognition. DC, USA, Washington
LeCun Y (2015) Learning methods for generic object recognition with invariance to pose and lighting. In: Computer vision and pattern recognition, 2004. CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on, vol 2. pp II–104). IEEE.
Li J, Bioucas Dias JM, Plaza A (2013) Spectral spatial classification of hyperspectral data using loopy belief propagation and active learning. IEEE Trans Geosci Remote Sens 51:844–856
Li W, Fu H, Yu L, Cracknell A (2016) Deep learning based oil palm tree detection and counting for high-resolution remote sensing images. Remote Sens 9:22
Li X, Wu T, Liu K, Li Y, Zhang L (2016) Evaluation of the Chinese fine spatial resolution hyperspectral satellite TianGong-1 in urban land-cover classification. Remote Sens 8:438
Li W, Wu G, Zhang F, Du Q (2017) Hyperspectral image classification using deep pixel-pair features. IEEE Trans Geosci Remote Sens 55:844–853
Lyu MR, Song J, Cai M (2005) A comprehensive method for multilingual video text detection, localization, and extraction. IEEE Trans Circuits Syst Video Technol 15:243–255
Ma Y, Wang L, Liu D, Liu P, Wang J, Tao J (2012) Generic parallel programming for massive remote sensing data processing. In: IEEE international conference on cluster computing, Beijing, China
Makantasis K, Karantzalos K, Doulamis A, Doulamis N (2015) Deep supervised learning for hyperspectral data classification through convolutional neural networks. In: Geoscience and remote sensing symposium (IGARSS), 2015 IEEE international. Italy, Milan
Marmanis D, Wegner JD, Galliani S, Schindler K, Datcu M, Stilla U (2016) Semantic segmentation of aerial images with an ensemble of CNSS. ISPRS Ann Photogramm Remote Sens Spat Inf Sci 3:473–480
Melgani F, Bruzzone L (2004) Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans Geosci Remote Sens 42:1778–1790
Mhaskar H, Liao Q, Poggio TA (2017) When and why are deep networks better than shallow ones? In: AAAI, pp 2343–2349
Plaza AJ (2009) Special issue on architectures and techniques for real-time processing of remotely sensed images. J Real Time Image Process 4:191–193
Saprykin O, Fedoseev A, Mikheeva T (2016) Recognition of urban transport infrastructure objects via hyperspectral images. In: VEHITS, pp 203–208
Schuster M, Paliwal KK (1997) Bidirectional recurrent neural networks. IEEE Trans Signal Process 45(11):2673–2681
Wu F-Y, Yan S-Y, Smith JS, Zhang B-L (2017) Traffic scene recognition based on deep CNN and VLAD spatial pyramids. arXiv:1707.07411
Zhang T (2001) An introduction to support vector machines and other kernel-based learning methods. AI Mag 22(2):103
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Hussain, S.A., Tahir, A., Khan, J.A. et al. Pixel-Based Classification of Hyperspectral Images Using Convolutional Neural Networks. PFG 87, 33–45 (2019). https://doi.org/10.1007/s41064-019-00066-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s41064-019-00066-z